Earlier this month, Just Security published a guest post by Sahr Muhammedally outlining a new congressional statute whereby the Defense Department can now move ahead in establishing a program that gives amends to foreign civilians who are injured as a result of US combat operations. The new law — section 8127 of the Consolidated Appropriations Act of 2014 — appropriates funds even for injuries that result from completely lawful US military actions. The idea is to respect the dignity of innocent war victims and provide at least some gesture of our concern for their loss. The architects of the program champion the ethical and strategic values of this effort, and, indeed, it appears to be the result in part of extraordinary work of organizations such as the Center for Civilians in Conflict.

The law, however, has a defect that could potentially undermine its own objectives. It allows payments to be made only if the foreign civilian is determined to be “friendly to the United States.” The law states:

“An ex gratia payment under this section may be provided only if … the prospective foreign civilian recipient is determined by the local military commander to be friendly to the United States”

The United States has a long history of making payments to foreign civilians for harms suffered as a result of US military actions. In World War II, the military found that such payments “created a wholesome respect for our Government and had a pronounced stabilizing effect.” Another instructive example was President Reagan’s actions after the downing of an Iranian civilian airliner by USS Vincennes. Reagan stated unequivocally that the US crew acted lawfully on the basis of a reasonable, though ultimately mistaken, assessment that the aircraft was an Iranian warplane. At the same time, the White House quickly announced that, “prompted by the humanitarian traditions of our nation, the President has decided that the United States will offer compensation on an ex gratia basis to the families of the victims,” and the administration explained that this gesture “is consistent with international practice and is a humanitarian effort to ease [the families’] hardship.” The President’s response was ethically and strategically wise.

Now imagine if Reagan had instead announced that the payments would be made only to those passengers who were determined to be “friendly to the United States.” It would have been a PR fiasco.

There are countless episodes like these in which one might envision the new amends law applying. It will apply not only in cases of mistaken identity such as the Iranian airliner. It will also apply in cases in which US forces conduct a lawful attack that knowingly kills some civilians, but meets the standard of proportionality due to the importance of the military target. And it will apply to less deliberate actions, such as a tank accidentally knocking over a city bus on its way into battle.

A virtue of the new amends law is that it is designed to replace an existing ad hoc system with a well-administered, permanent program that still allows for significant commander discretion—that is, except for this eligibility requirement.

So what exactly might it mean for a commander to have to determine a civilian was “friendly to the United States,” and who could be turned away from the program? How about a reporter who once published a propaganda story hostile to the United States? Or a civilian who refused to inform US forces about the whereabouts of her family member who is a suspected Taliban member? Or how about someone who is neutral rather than “friendly,” such as a local villager who does not even have an opinion about the United States of America? Or a homeless person about whom we know little else? When a group of civilians is killed in a single incident, will a commander have to sort through their files to determine which ones fall into the “friendly” category?

I worry that categorically excluding this class of civilians (i.e., disallowing discretion for US military commanders) could undercut the larger goal of stabilizing relations with foreign populations. And, indeed, in some cases the new law might be worse than the prior ad hoc system in which US commanders appear to have had greater discretion along this dimension.

So where did this requirement come from? Congress appears to have imported the language verbatim from the Foreign Claims Act, which provides compensation, under particular conditions, to foreigners injured as a result of noncombat operations. That Act also limits payments to individuals who are “friendly to the United States.” A claim may be allowed only if:

“in the case of a national of a country at war with the United States, or of any ally of that country, the claimant is determined by the commission or by the local military commander to be friendly to the United States”

But that clause applies only to wars with other states, and not to citizens of friendly host countries. The rationale for the clause is thus not directly applicable to transnational armed conflicts in which the US is presently engaged.

Perhaps more fundamentally, the Foreign Claims Act was drafted against a background in which the prevailing humanitarian norm allowed governments to largely disregard the interests of “enemy nationals” – citizens of the opposing state. Understood in that historical context, the “friendly to the United States” clause in the Foreign Claims Act was progressive in its conception. Rather than exclude all enemy nationals, the government extended benefits to enemy nationals who favored the United States.

The “friendly to the United States” proviso in the new amends law, however, may be understood to have an opposite effect. It may well send a message to populations in Afghanistan and elsewhere that the US cares about the loss of civilian life only when the dead and injured can be proven to have supported, liked or sided with us, and it will suggest that the US government’s baseline assumption is they don’t.

One way to try to address this defect in the law is in the implementation of the statute. Perhaps the exclusion could be interpreted narrowly, or regulations could set a presumption in favor of friendliness unless proven otherwise. But those are not the most ethical and strategically wise options. Nor would they be truly effective. The best outcome may be simply to revise the amends law and drop or qualify this exclusion.